Commit 11aff324 authored by pengdonglin's avatar pengdonglin Committed by Steven Rostedt (Google)
Browse files

ftrace: Fix preemption accounting for stacktrace filter command 

The preemption count of the stacktrace filter command to trace ksys_read
is consistently incorrect:

$ echo ksys_read:stacktrace > set_ftrace_filter

   <...>-453     [004] ...1.    38.308956: <stack trace>
=> ksys_read
=> do_syscall_64
=> entry_SYSCALL_64_after_hwframe

The root cause is that the trace framework disables preemption when
invoking the filter command callback in function_trace_probe_call:

   preempt_disable_notrace();
   probe_ops->func(ip, parent_ip, probe_opsbe->tr, probe_ops, probe->data);
   preempt_enable_notrace();

Use tracing_gen_ctx_dec() to account for the preempt_disable_notrace(),
which will output the correct preemption count:

$ echo ksys_read:stacktrace > set_ftrace_filter

   <...>-410     [006] .....    31.420396: <stack trace>
=> ksys_read
=> do_syscall_64
=> entry_SYSCALL_64_after_hwframe

Cc: stable@vger.kernel.org
Fixes: 36590c50 ("tracing: Merge irqflags + preempt counter.")
Link: https://lore.kernel.org/20250512094246.1167956-2-dolinux.peng@gmail.com


Signed-off-by: default avatarpengdonglin <dolinux.peng@gmail.com>
Signed-off-by: default avatarSteven Rostedt (Google) <rostedt@goodmis.org>
parent e3333326

kernel/trace/trace_functions.c

+1 −5
Original line number Diff line number Diff line
@@ -633,11 +633,7 @@ ftrace_traceoff(unsigned long ip, unsigned long parent_ip, 


static __always_inline void trace_stack(struct trace_array *tr)

{

	unsigned int trace_ctx;



	trace_ctx = tracing_gen_ctx();



	__trace_stack(tr, trace_ctx, FTRACE_STACK_SKIP);

	__trace_stack(tr, tracing_gen_ctx_dec(), FTRACE_STACK_SKIP);

}



static void